South-eastern Pashto: Khost, Paktia and the Karlanri speech

At a glance

• The Linguistic Data Consortium's four-way grouping of Pashto speech has a south-eastern group centred on Quetta. It names the Sherani, Bannu and the north and south Waziri in it, and gives no group to Khost or Paktia. [6]

• The speech of the Karlanri tribes (Waziri, Bannuchi, and the Zadran dialect of Khost and Paktia) is set apart less by its consonants than by a shift of vowels: ā, o and u become o, e and i. [2, 3, 7]

• Herbert Penzl wrote in 1959 that in the dialect of Khost the standard words plār "father", mor "mother" and lur "daughter" are said ploor, meer and liir. [4]

• How ښ and ږ are said there has more than one answer. The Zadran of Khost and Paktia have palatal sounds, Waziri and Bannuchi a plain sh and zh, and the sources do not agree on every point. [6, 7]

• Georg Morgenstierne noted that the tribes whose dialects show the vowel shift are, in Pashtun tradition, of a different genealogical line from the rest of the Afghan tribes. [2]

What south-eastern means

"South-eastern" is the label the Linguistic Data Consortium uses for one of its four dialect regions, and it is a label for sounds. The region is the one whose nearest large city is Quetta, and the report lists the southern parts of the North-West Frontier Province, the Sherani, Bannu, the Waziri in their northern and southern groups, Balochistan, the Kakar, Achakzai and Tarin. [6] In that group ښ and ږ are a plain sh and zh, so the language is Pashto there, as it is at Quetta. [6] The general picture is in Pashto dialects: soft and hard speech.

The same report names the Mangal among the main Pashto-speaking tribes, but it places no one from Khost or Paktia in any of its four groups. [6] The speech of Khost, Paktia, the Kurram valley and the hills towards Bannu and Waziristan is better described by a second line of division that cuts across soft and hard speech.

Two ways of dividing the dialects

Morgenstierne listed the main points on which Pashto dialects differ in the report of his linguistic mission to North-Western India. The first was the old contrast between soft dialects, which keep the sounds ṣ̌ and ẓ̌ or turn them into sh and zh, and hard north-eastern ones, which make them kh and g. [2] The second was a change in the vowels. In the east-central dialects, he wrote, from Afridi to Waziri, there is a tendency to change the vowel system, and in the Waziri and Bannu dialects where it is carried furthest, ordinary Pashto ā, o and u become o, e and i. [2]

The vowel change has spread from that centre to hard dialects too. Morgenstierne named Afridi and Bangash. [2] In his later article for the Encyclopaedia Iranica he put the consequence plainly: dialects with a further development of the common vowels may belong either to the soft or the hard group. [7] The Karlanri speech is therefore not one more point on the soft-to-hard line. It runs across it.

D. N. MacKenzie, writing in 1959, held that the morphological differences between the extreme north-eastern and south-western dialects are few and unimportant and that the criteria of dialect differentiation in Pashto are primarily phonological. [5] Penzl, in the same year, found that syntax differs little between dialects, and that what differs is vocabulary, endings and speech sounds. [4]

The Karlanri vowel shift

Morgenstierne remarked that the tribes whose dialects show the shift are, according to tradition, of a different genealogical descent from the rest of the Afghan tribes. [2] Ibrahim Khan, a barrister writing in the Pashto Academy's journal Pashto in 2021, calls them the Karlāṇ tribes and the change "the Great Karlāṇ Vowel Shift". He sets out what the linguists found and adds recordings of speakers from Bannu, Waziristan and Khost. [3]

Khan quotes Corey Miller's description of the three vowels involved: /ā/, /o/ and /u/, "which in M can shift to /o/, /e/ and /i/, respectively". [3] In the Iranica article Morgenstierne gives the detail. Stressed o moves towards ȯ, ȫ or ɛ, so that Waziri has mɛ̄r for mor, "mother"; ā rounds towards â or å; and in Waziri ū becomes ī, so līr is "daughter". [7] The same article records that Kakari of Shahrig in Balochistan has the Waziri-type forms līr, mēr and calēr, "four". [7]

Two words from the recordings in Khan's article show what the change sounds like. A 78-year-old Bannu speaker, recorded by Yousaf Khan Jazab, says "father" as پلور plor, with the long rounded vowel /ɔː/. A Marwat speaker says the same word as پلار plār, with the vowel of Kandahar and Yusufzai. [3] Khan's reading of the recording is that Elfenbein was right to say that Bannuchi mainly goes with northern Waziri. [3]

Khost shows the shift in the same words. Penzl, who worked on Pashto dialects in Afghanistan in 1958 and 1959, wrote that in the dialect of Khost in Paktia ploor is said for standard plār, meer for moor and liir for lur. [4] Those three words carry the three changes: ā to o, o to e and u to i. Penzl called them extreme types, and added that such types are "not acceptable" as a spoken standard. [4]

Where the shift stops

Not every Karlanri tribe has it. Khan reports that the Wardak, a northern Karlanri tribe by his reckoning, have no regular vowel shift, though they say ā as a rounded å, and that even among the Orakzai, whose speech is touched by it, ā becomes o and no more. [3] The earliest outside notice he found is in the Bannu district gazetteer of 1883-84, which says that "Waziris, Bannúchís, Khataks and Marwats each pronounce the vowels somewhat differently and make use of many words peculiar only to themselves". [3] Khan notes that the gazetteer wrongly included the Marwat. [3]

The Zadran of Khost and Paktia

The best description of Khost speech is Daniel Septfonds's Le dzadrâni: un parler pashto du Paktyâ (1994). Khan, who draws on it, reports that Septfonds collected his material in two districts, Tani and Spera, now in Khost province, which was formed out of old Paktia. [3] His main informant, Shan, came from Ḍaṇḍkay village east of Tani. [3]

Khan quotes Anna Boyle David's account of the sounds. In Dzadrani, she writes, ښ is pronounced as a palatal /ç/ and ږ as a palatal /ʝ/. [3] That is neither the retroflex of Kandahar nor the kh and g of Peshawar. Khan hears it in a recording of a Khost speaker and writes it x̌ and ǵ, as in mīǵ, "we", and x̌ə, "good". [3] His summary table at the start of the article, however, lists Dzadrani with Afridi among the dialects with velar x and the vowel shift. [3] The table and his transcriptions give two different sounds for the same speech.

One verb shows the Zadran speech clearly. Septfonds records drim- as the present stem of the verb "to go", which standard Pashto forms from other stems. Khan found the same stem in the Karlanri dialects and in Tarino (Wanetsi). [3] Septfonds's informant Shan gave this sentence, with the aorist stem in the first command and drim- in the second:

ته که چېرې په قول وې، خو وې سره لوړشه؛ که په قول نه وې، بیا مه دریمه

Tə ka čere pə kawl we, xo we sra loṛ-ša. Ka pə kawl nə we, xo bya ma drima.

If you are a man of your word, then go with them. If you are not, then do not go.

Khan prints Septfonds's French ("Si jamais tu es de parole, alors pars avec eux. Si tu ne te sens pas de parole, alors ne t'en va pas!") and his own English. [3] The English above follows the French.

Khan adds a sentence from a Khost speaker, taken from a Zhwandoon TV programme:

اغه رېغ مول د میږ چاینه ته دریمي

Áğa reğ mol də mīǵ čāina tə drimī

All those goods of ours go to China.

Here drimī is again the stem for "go", and mīǵ the palatal "we" in place of the standard mūnğ. [3]

Consonants in the south-east and Paktia

Two consonant changes are recorded outside the Zadran area. Penzl noted that in many dialects of the provinces of Kabul and Paktia, as in Ghazni, Kandahar, Herat and Dera Ismail Khan, z appears instead of zh in some very common words. The Pashto Academy meeting of 1942 rejected the z-forms, and in 1958 the meeting warned against writing z or j for zh, as in zeba or jeba for zheba, "tongue, language". [4]

Morgenstierne noted that Bannuchi turns the affricates c and j into s and z, as many north-eastern dialects do, and that this tends to remove sounds foreign to the neighbouring Indo-Aryan languages. [2] The Linguistic Data Consortium's south-eastern group keeps them as ts and dz. [6] The two accounts differ about Bannu, and the older one is a single observation from a short visit.

Lorimer, a British political officer who compiled a Waziri grammar in 1902, glosses lwar in his vocabulary as "high" and also "hard, harsh", and gives the example sentence that the Waziri dialect is hard and that of the Bannuchis soft. [1] Bannu is in the Linguistic Data Consortium's group that says ښ as sh. [6]

Words and endings

The grammar of the south-east differs from standard Pashto in small things. Khan quotes Jazab on the first person singular of verbs: the nasal ə ending of Bannu, Waziri and Khattak is "a prominent feature of the Karlanri varieties", and no other variety has it except Marwat, which borders them. [3] Where Kandahar and Yusufzai say "I want" with an ending in -m, Bannu ends it with ə. [3]

Some words differ too. Khan notes that Karlanri and Marwat use the stem kəs- for "to look" where Yusufzai and Kandahar use gor-. [3] Penzl noted ster for "great, big", which he thought was said only in southern and south-eastern Paktia and parts of Bannu and Kohat, and which appeared in Kabul newspapers beside the common looy. [4] Morgenstierne found Jaji pēṇ- beside the usual pinjə, "five". [7]

Waziri has its own vowels in everyday speech. Khan, who hears the word for "rain" as wör, with a front rounded vowel, gives the sentence وېر وریږي, Wör warī, "It is raining". [3] He adds that the Pashto alphabet "is not suited to the intricacies of Waziri vowels", and asks readers to look at his transliteration. [3] The Waziri differences from Peshawar speech, and Lorimer's examples, are set out in Pashto dialects: soft and hard speech.

Where the speech is placed

Pashto is written with one spelling for all these speakers. Penzl observed that speakers of dialects whose sounds differ from Kandahar's still accept the Kandahar orthography as standard when they write. [4] The Pashto Academy meetings in Kabul in 1942, 1948 and 1958 dealt only with spelling, and no spoken standard, he wrote, has ever been formulated, described or taught. [4]

Khan's article asks whether the Karlanri dialects and the distant Tarino of Balochistan are connected, and gives about fifty words that his Tarino informant noticed on a visit to Bannu. [3] Morgenstierne thought the Karlanri vowel change was not very old: most Pashto loanwords in Ormuri have the ordinary Pashto vowels, and not those of the neighbouring Waziri. [2]

The surrounding country is covered in Bannu, Waziristan, the Kurram valley and Spin Ghar, and the wider language in Pashto: language family and dialect variation. The poet Pir Muhammad Karwan is from Tani in Khost.